A Knowledge Management System (KMS) Using a Storytelling-based Approach to Collect Tacit Knowledge

149  Download (0)

Full text


Nova Southeastern University


CEC Theses and Dissertations College of Engineering and Computing


A Knowledge Management System (KMS) Using a

Storytelling-based Approach to Collect Tacit


Nicholas Shaw

Nova Southeastern University,doc@docharley.com

This document is a product of extensive research conducted at the Nova Southeastern UniversityCollege of Engineering and Computing. For more information on research and degree programs at the NSU College of Engineering and Computing, please clickhere.

Follow this and additional works at:https://nsuworks.nova.edu/gscis_etd Part of theComputer Sciences Commons

Share Feedback About This Item

This Dissertation is brought to you by the College of Engineering and Computing at NSUWorks. It has been accepted for inclusion in CEC Theses and Dissertations by an authorized administrator of NSUWorks. For more information, please contactnsuworks@nova.edu.

NSUWorks Citation

Nicholas Shaw. 2018.A Knowledge Management System (KMS) Using a Storytelling-based Approach to Collect Tacit Knowledge.Doctoral dissertation. Nova Southeastern University. Retrieved from NSUWorks, College of Engineering and Computing. (1033)


A Knowledge Management System (KMS) Using a Storytelling-based

Approach to Collect Tacit Knowledge


Nicholas James Shaw

A dissertation submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy


Computer Information Systems

College of Engineering and Computing Nova Southeastern University



An Abstract of a Dissertation Submitted to Nova Southeastern University In Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy A Knowledge Management System (KMS) Using a Storytelling-based Approach to

Collect Tacit Knowledge By

Nicholas J. Shaw December 2017

Since the 1990s, Knowledge Management Systems (KMS) have been largely unsuccessful in the collection of tacit knowledge. The process, whether through direct input by the holder of the tacit knowledge or through an intermediary such as the

collection of tacit knowledge through interviews and videos, has not succeeded. Reasons encompass the organizational (such as culture of the organization), the technological (example: poor tools), and the individual (example: knowledge is power, i.e. where experts with rare knowledge results in knowledge hoarding instead of transfer). The purpose of this study was to demonstrate that tacit knowledge could be successfully and consistently collected from the participants themselves and placed into a KMS using a storytelling-based approach. This study extended past research that collected stories for KMS’ using interviews and videos by having participants directly entering their data, as stories, into a KMS. This was a new approach and it was posited that having participants use stories to enter their tacit knowledge themselves into a KMS would overcome their reluctance to provide tacit knowledge thus overcoming barriers to providing tacit knowledge into a KMS.

The validation methodology was based upon three elements: the deep-dive research element, the issues and solution element, and the dissertation proposition element. The deep-dive research element was the extensive research for the study into knowledge management, storytelling, and other various methods for collection of tacit knowledge. The issues and solution element consisted of issues about tacit knowledge that were identified from the deep-dive research element, i.e. general arguments

constructed about knowledge management which were backed by data from research into knowledge management systems and storytelling. Theoretical solutions to the issues regarding the capture of tacit knowledge were then constructed which included the storytelling-based approach and a KMS framework for the collection of tacit knowledge. Lastly was the dissertation proposition element which consisted of a thorough analysis of the survey data against each of the dissertation propositions. There were three

propositions. Proposition 1 was sharing of knowledge and the storytelling-based

approach. Proposition 2 was about the framework, the scenarios, guiding questions, and Communities of Practice (CoP), and Proposition 3 was about participant knowledge and interaction with forums. Each proposition was evaluated independently.


The study was successful and validated propositions 1 and 2. For proposition 1, 81% of the participants responded positively to the eight study questions directed towards this proposition. For all eight questions across all 21 participants, the mean was 29.952 against a target test mean of 24 with a range of 27.538-32.367. For proposition 2, 76.19% of participants scored this section positive. For all six questions across all 21 participants, the mean was 23 against a target test mean of 18 with a range of 21.394-24.606. However, the results for proposition 3 were inconclusive and must be considered a failure. Most of the respondents either scored ‘no change’ to at least 50% of the

questions or they stated they had never been to a forum. For all four questions across all 21 participants, the mean was 12.905 against a target mean of 12 with a range of 11.896-13.914. Based upon propositions 1 and 2, the null hypothesis was disproved.

Participants liked the storytelling-based approach, providing their tacit knowledge, and they liked the framework.



I consider it a privilege to have completed this dissertation under the guidance of three of the finest individuals and scholars that I have ever had the pleasure of working with.

First, I want to offer my sincere thanks and gratitude to Dr. Peixiang Liu, my committee chair and professor, who literally taught me the meaning of good research through his wisdom, his feedback, and his advice on other topics that I believed to be important. He helped me get a conference paper published through the IEEE on the topic of my dissertation.

Dr. Gurvirender Tejay, one of my professors and committee members, who always helped improve the content of my dissertation with his insightful input and excellent suggestions.

Dr. James Cannady, one of my professors and committee members, for his excellent advice for necessary changes in my dissertation.

Mrs. Lisa Shaw, my loving wife, who encouraged me to get another Doctorate and then put up with all of my long hours in the course work and working on this dissertation. And she still managed to give me her love and encouragement. She truly understands the level of effort and commitment a Doctoral program requires from everyone; I could not have done this without her support.

I would also like to thank Mr. Umesh Phuyal who developed the web-based knowledge management system that was used to conduct the experiments. I had spent a great deal of time learning the PHP programming language, HTML 5 programming, and the MySQL relational database but was still falling behind. I reached out to Mr. Phuyal and he had the system up-and-running in a few weeks.

Finally, I want to express my gratitude to the entire faculty and staff who have made my studies and research at Nova Southeastern University one of the best


Table of Contents

Abstract 2 Table of Contents 5 List of Tables 8 List of Figures 9 Chapter 1: Introduction 10 Background 10 Dissertation Goal 18 Problem Statement 20 Research Questions 23

Relevance and Significance 24 Barriers and Issues 26

Assumptions, Limitations and Delimitations 29 Limitations 29

Delimitations 29

Definition of Terms 30

Summary 33

Chapter 2: Review of the Literature 36 Knowledge Sharing 36

Knowledge Management Systems 37 Forums 40

KMS Approaches 41

KMS Architectures 41

Database Management Systems 41 Case-Based Reasoning (CBR) 42 Ontology-based KMS’ 42

Storytelling 42

Summary 45

Chapter 3: Methodology 48

Overview of the Quantitative Approach 52

Overview of the Qualitative Approach 57

Design and Implementation of the KMS 58


Physical host: 68 Virtual Machine: 68 Success Criteria 70 Chapter 4 Results 79 Data Analysis 79 Findings 83 Framework 83

Means to Solicit Participants 84

Participation 84

Summary of Results 88

Chapter 5 Conclusions, Implications, and Recommendations 90 Conclusions 90

Scientific Research Contribution 92

Implications 92

Limitations/Recommendations 93

Summary 95 Appendices 96

Appendix A Waiver of Documentation of Informed Consent 98

Appendix B Study Information Sheet 102

Appendix C Participant Data and Survey 107

Appendix D Story Guiding Questions 114 DOMAIN: Migration 114

Subdomain: Application and/or Database 114 Subdomain: Data Center Migration 115 Subdomain: Servers 116

Subdomain: Storage 116

Subdomain: System Software 117

DOMAIN: Provisioning 117 Subdomain: Cloud 117 Subdomain: Networks 118 Subdomain: Servers 118 Subdomain: Storage 119 Subdomain: Virtualization 119

DOMAIN: Design System 119

Subdomain: Hardware/Software 119

DOMAIN: Decommission 120 Subdomain: Application 120


Subdomain: Servers 120 Subdomain: Storage 121

Appendix E Actual Story Example 122

Appendix F Participant Response Scoring 123

Appendix G KMS Screen Captures 126

Appendix H Survey Code Book 138


List of Tables


Table 1 - IT Domains/Subdomains utilized within the study 67

Table 2 – Example of an Unsuccessful Research Hypothesis (𝒉𝒂 ) for Proposition 1 73

Table 3 - Participant Age Breakdown 85

Table 4 - Participant Region Breakdown 85

Table 5 - Summary of Results 86

Table 6 - Participant Survey Response Table 123

Table 7 – Demographics & Participant Status 124


List of Figures


Figure 1 - Information systems are interrelated systems of technical and organizational elements 19

Figure 2 - Use Cases for KMS 53

Figure 3- End-to-End Participant Process 53

Figure 4 - Overall System Architecture 61

Figure 5 - KMS Schema Model 62

Figure 6 - Guiding Questions Table Structures 65

Figure 7 – Minitab 17 Sample t Test for the Mean of Proposition 1 Summary Report 81

Figure 8 - Minitab 17 Sample t Test for the Mean of Proposition 2 Summary Report 81

Figure 9 - Minitab 17 Sample t Test for the Mean of Proposition 3 Summary Report 82

Figure 10 - Minitab 17 Sample t Test Diagnostic Report for the Mean of Proposition 1 87

Figure 11 -Minitab 17 Sample t Test Diagnostic Report for the Mean of Proposition 2 87

Figure 12 - Minitab 17 Sample t Test Diagnostic Report for the Mean of Proposition 3 88

Figure 13 - KMS Home Screen 126

Figure 14 - Contributor Password Screen 127

Figure 15 - Contributor Initial Screen 128

Figure 16 - Contributor Status/Edit/View Stories Screen Capture 129

Figure 17 - Selecting a domain and subdomain pair of a story 130

Figure 18 - Answering guiding questions screen capture 131

Figure 19 - Continuing the creation of a story screen shot 132

Figure 20 - Lessons learned screen shot 133

Figure 21 - Collaboration between contributor and reviewer screen shot 134

Figure 22 - Completed story being reviewed screen shot 135

Figure 23 - Reviewer Accept or Decline Story screen shot 136

Figure 24 - Logging out of the KMOS system 137




Knowledge management systems (KMS) refer to any kind of Information Technology (IT) system that stores and retrieves knowledge, improves collaboration, locates knowledge sources, mines repositories for hidden knowledge, captures and uses knowledge, or in some other way enhances the knowledge management (KM) process (Becerra-Fernandez, 2000; Frost, 2013; Jimenez-Jimenez, Martinez-Costa, & Sanz-Valle, 2014; Rance & Hanna, 2007). Chen, Xiao, Ren, and Shi (2011) defined knowledge management (KM) as the process that enterprises use to identify and organize knowledge and then effectively use the knowledge to competitive advantage. Fanfan (2012)

described KM as any process or practice of creating, acquiring, capturing, sharing, using and evaluating knowledge wherever it resides.

There are two fundamental types of knowledge – explicit and tacit. According to the Cambridge dictionary, explicit knowledge is knowledge that can be expressed in words, numbers, symbols, and stored. The Law Dictionary states it is knowledge that is recorded and expressed; it is easy to share and store and is the opposite of tacit

knowledge. Tacit knowledge, according to the Cambridge Dictionary, is knowledge you get from personal experience. The Law Dictionary states that tacit knowledge is

unspoken, unwritten, and hidden stores of knowledge based on experiences, emotions, institutions, insights, and observations. Examples of tacit knowledge include the knowledge of how to ride a bicycle, how to knead bread, and how to use a word processor (Linde, 2001). In short, tacit knowledge is the knowledge that resides in our heads and is far more difficult to represent in a knowledge management system (KMS)


11 due to the reluctance of the owners of the knowledge to allow it to be placed into a KMS (R. O. Weber, 2007). Researchers who have attempted to populate KMS’ with tacit knowledge concede that there are two fundamental obstacles – fear (fear among

employees that sharing knowledge reduces job security) and power (keeping information to oneself enhances job security) (Disterer, 2001; R. O. Weber, 2007). The power factor is, stated simply, knowledge is power and keeping that knowledge to oneself contributes to organizations retaining them and forces people to come to them for their knowledge contributing to self-worth (Benbya & Alstyne, 2008; Fanfan, 2012; Kankanhalli, Tan, & Wei, 2005). Other researchers are not convinced that either fear or power is a primary factor. Okoroji, Velu, and Sekaran (2014) determined that the key factor was appropriate motivation. Okoroji, et al. did note that in every organization, there are individuals who are willing to share their knowledge, and there are those who prefer not to. Their research sought to understand why some are willing while others are not, i.e. what initiatives encourage knowledge sharing and what are the barriers to knowledge sharing. Riege (2005) suggested there are three primary barriers – individual, organizational, and technological. According to Riege, the individual barriers include lack of communication skills, lack of social networks, differences in culture, lack of time, lack of trust, lack of motivation, and fear of not receiving recognition. He stated that “knowledge sharing practices often seem to fail because companies attempt to adjust their organizational culture to fit their KM, instead of implementing knowledge sharing practices that fit their culture” and the technology (hardware and software tools) necessary to implement successful solutions; Riege included in this a shortage of appropriate software tools. Ling, Sandhu, and Jain (2009) supported Riege but their research was limited to the opinions of executives; no individual participants were part of their study. Sandhu, Jain, and Ahmad (2011) picked up where Ling, Sandhu, and Jain left off and this time went


12 technological barriers do exist.

This researcher asserts that most domain experts do share their tacit knowledge willingly with others on a day-to-day basis. They share their knowledge with junior members who are learning their craft, they share their knowledge with other domain experts as they recount lessons they have learned over time or from specific incidents and/or challenges and they share their tacit knowledge with management in briefings or as concerns to management or other domain experts. This is supported by Guechtouli, Rouchier, and Orillard (2012) and Ariffin, Arshad, Shaarani, and Shah (2007) in their discussions on direct knowledge transfer and by Sandhu, Jain, and Ahmad (2011) in their research into knowledge sharing. Based upon the three barriers of Riege, this research targeted the barriers of individual and technology.

Early KMS efforts in the 1990s to capture tacit knowledge of employees were geared towards employees ready to retire or leaving the organization on their own for other employment opportunities. The concerns were the loss of valuable lessons learned while at the organization. The research of Benbya and Alstyne (2008) at HP and Siemens demonstrated that fear can be overcome with the right motivation. This supports the issue of motivation discussed in Okoroji, Velu, and Sekaran (2014). HP developed an incentive program based on frequent flyer mile certificates; however, after 90 days, only 20% of the target audience had participated. HP elected to continue the program without change, primarily due to the large number of mile certificates they still had on hand. Siemens, another global company, took a different approach by rewarding the country that produced the knowledge and to the countries that consumed it and the countries rewarded the individuals. Siemens rewarded both producers and consumers of the


13 knowledge with corporate stock shares and the financial value of the shares were based upon the type of contribution. However, over time, managers found it difficult to continue the incentive program as those who received the financial shares were not turning them in for the money but keeping them. This was still considered a success as those who received the shares considered the number of shares a badge of sorts, i.e. the more you had, the greater was your importance.

Weber (2007) identified nine reasons why KMS' may fail. In evaluating these, this researcher determined that three are relevant to this study because they relate in one way or another to the collection or failure to collect tacit knowledge (see below).

Examples of non-applicable reasons were: KM approaches may fail when they attempt to create a monolithic organizational memory and another was KM approaches may fail when they are outside the process context. The following reasons were applicable to this study:

• KMS' often fail due to the nature of the KMS, i.e. there are no bounds on what a domain expert can enter or how.

• KMS’ may fail when users are afraid of the consequences of their contributions; this is related to job security. Users may even feel that withholding their knowledge may be a way to secure influence.

• KMS’ may fail when they do not integrate humans, processes, and technology In the context of Riege’s three barriers (individual, organizational, and technological), Weber’s reasons fall within the technological and individual.

It is in the area of tacit knowledge collection that KMS' have not been successful. Specifically, as noted earlier, for reasons that span individual, organization, and


14 2014, researchers considered tacit knowledge as not being able to be codified (Qiu, Want, & Nian, 2014). In 2015, Rumanti, Hidayat, and Suputro (2015) stated that tacit

knowledge is the most difficult to transfer to others and in 2017, Patalas-Maliszewska, Krebs, and Dudek (2017) stated that tacit knowledge is difficult to attain in KMS’; however, Pratt-Whitney Rocketdyne’s (PWR) Goldfire KMS had some success (Chun, Sohn, Arling, & Granados, 2008). Goldfire was essentially a two-part implementation where the first part, the AskMe portion, was fundamentally a forum where individuals identified themselves as experts in one or more areas and people could present questions to them. This encouraged experts to share their knowledge. The second part, the

Goldfire KMS, searched for data throughout the many KMS’ or knowledge within PWR. Thus, the Goldfire KMS became the single source of information. While not all

information resided within Goldfire, Goldfire was able to search out and find information throughout the many repositories within PWR.

In addition to types of knowledge, there are two types of knowledge transfer - direct and indirect (Guechtouli et al., 2012). Direct knowledge transfer is one-on-one, one-on-many, or many-to-many (such as a meeting) but face-to-face such as mentoring or coaching. Ariffin, Arshad, Shaarani, and Shah (2007) used the example of a domain expert (DE) guiding a novice user through a procedure - the DE transfers their tacit knowledge to the user and, in this case, the tacit knowledge is used to improve the work activities of the novice user. In this example, the DE is using direct communications (personalization) and direct knowledge transfer versus a tool. Indirect knowledge transfer can be through any means where different people at different times can view the


15 artifacts of the acquired knowledge. Examples of indirect knowledge are written

documents, videos, forums and KMS'.

Companies have created internal Wiki's for knowledge sharing and there are forums on almost any topic where knowledge sharing takes place as well as within other social media such as Facebook (indirect knowledge transfer). On any given day, there are knowledgeable employees sharing their knowledge with less knowledgeable employees to help them learn (direct knowledge transfer). This was a finding of the Goldfire team at PWR (Chun et al., 2008) as well as studies on direct and indirect knowledge transfer referenced earlier. It is supported by Okoroji, Velu, and Sekaran (2014) and Sandhu, Jain, and Ahmad (2011). Thus, the transfer of tacit knowledge does occur.

Wasko and Faraj (2005) researched electronic forums into why participants participated since there is no immediate benefit to them and free-riders are able to acquire the same knowledge as everyone else. The research of Wasko and Faraj showed that those seeking knowledge have no control over who responds to their questions or the quality of the responses and participants have no assurances that those they are helping will ever return the favor. The researchers concluded that individuals contribute knowledge to electronic media when they perceive that it enhances their professional reputations, and to some extent, it is enjoyable to help others. Individuals who contribute knowledge do not seem to expect help in return.

The collection of tacit knowledge into KMS' was the point of research by Coffey and Hoffman (2003) who tied the collection of tacit knowledge to the organizational need to retain institutional knowledge in order to advance the mission of the organization, avoid making the same mistakes over again and to leverage the accomplishment of


16 Benbya and Alstyne (2008) and others (Orth, Smolnik, & Jennex, 2009; Pumareja & Sikkel, 2005; Vizcaino, Soto, Portillo, & Piattini, 2007).

The focus of this research targeted the technological and individual obstacles as noted by Riege through the collection of tacit knowledge into a KMS by those who possess the knowledge using a storytelling-based approach. As seen up to this point, the transfer of tacit knowledge through direct and indirect methods does occur; however, the elicitation of tacit knowledge into a KMS by those who have the knowledge is still considered a major challenge. Whyte and Classen (2012), in the Journal of Knowledge Management did collect tacit knowledge for a KMS using stories; however, it was via face-to-face interviews with the data later inserted, by the researchers, into the KMS. This is not an efficient approach as it takes much longer to conduct the interviews and then to insert the data into a KMS as well as make updates. This study proposed having the DE contribute their tacit knowledge into the KMS themselves with the vehicle of elicitation being stories. The literature either says it cannot be done (Fanfan, 2012) or it is done through interviews and recordings. In 2001, Swap, Leonard, Shields, and Abrams (2001) explored storytelling to transfer knowledge in the workplace and, like Whyte and Classen in 2012, they used interviews to collect the tacit knowledge (as did Schank (2010)). This research proposed using the KMS itself to collect the tacit knowledge with no intermediary.

A storytelling-based approach is different in that it utilizes mechanisms employed in direct knowledge transfer. Storytelling has been used to record issues and lessons learned in project management (Buttler & Lukosch, 2012b), education (Sugathan & Kalid, 2009), requirements elicitation (Boulila, Hoffmann, & Herrmann, 2011) and


17 capturing tacit knowledge at MITRE (Kalid & Mahmud, 2008). Like earlier research using stories to collect tacit knowledge, the work at MITRE was through interviews (direct knowledge transfer).

The goal of this study was to determine if, based upon the research, the use of storytelling within a KMS, by the participant, could be successful in the collection of tacit knowledge. When individuals pass along their tacit knowledge to others, they often do it through stories. Thus, the right condition for the collection of tacit knowledge was the ability to tell a story using a KMS. Researchers across the social sciences as well as KMS researchers have noted that people love a good story (Linde, 2001; Schank, 2010; Sugathan & Kalid, 2009; Whyte & Classen, 2012). Through storytelling and a KMS, it was hypothesized that the barrier of technology and individual would be overcome.

People, telling verbal stories, often make assumptions about the listener that can result in lost details and misunderstanding to the reader or listener, i.e. the person telling the story assumes that the listener knows what they’re talking about. Even though they may be in the same field (Information Technology) and both are UNIX administrators, that assumption can easily be invalid. Assumptions could be as simple as one systems administrator talking to another systems administrator where one is talking IBM UNIX and its’ virtualization technology while the other is an HP UNIX administrator and the technologies are not the same so the HP UNIX administrator doesn’t understand.

Another example might be a systems administrator talking about RAID 10 but the second administrator knows RAID 1 (mirroring) and RAID 0 (stripping) but has never heard the term RAID 10 so doesn’t understand.

To create stories that are productive and meet a need, a framework must exist for telling the story. This study proposed a framework that consisted of the basic elements of a narrative story (Linde, 2001) and guiding questions that were asked based upon a


18 questions provide a frame of reference for the reader and a sort of fill-in-the-blanks to reduce the occurrence of assumptions. Using the RAID example above, the guiding questions might include a definition of RAID 10. Think of domains/subdomains as categories such as provisioning (domain) and servers (subdomain). The collected

information is then merged into a story. The basic elements of the knowledge framework were the separate components of the guiding questions, who, what, when, where, why, how, the impacts, the obstacles encountered, and the lessons learned. This aided in extracting the story from the DE and ensured that sufficient information was provided to the reader. Once the tacit knowledge was placed into the KMS, the tacit knowledge became explicit knowledge.

Dissertation Goal

The goal of the study was to demonstrate that the use of storytelling could be successful in the collection of tacit knowledge by participants who directly entered their tacit knowledge, through stories, into a KMS. It, thus, demonstrated a solution to the obstacle of technology and individual in knowledge sharing.

The research addressed the relevant issues identified by Weber (2007) in the problem statement using a ‘KM in the small’ based approach (Orth, Smolnik, & Jennex, 2009). It extended this to incorporate the application of Schank's (2010) and Whyte and Classen's (2012) storytelling-based approach using scenarios. Scenarios utilize

communities-of-practice (CoP) thus helping to reduce assumptions on the part of the participants regarding the reader. This approach integrates humans, processes, and technology (Figure 1), is intuitive to DE, is consistent, and it enhances knowledge sharing. According toKroenke (2011), information systems are interrelated systems of


19 technical and organizational elements. The technology is the hardware (network, servers, and storage) that supports the KMS (the software/tool) and the data (the information contained within the KMS). The approach is consistent from the standpoint that storytelling is a widely used mechanism to convey knowledge to others. DE foundthe KMS more intuitive to use based upon earlier research into stories and storytelling thus enhanced the sharing of their tacit knowledge with others through a structured database management KMS architecture. As noted earlier, Whyte and Classen (2012) felt storytelling to be the best way to transfer tacit knowledge and storytelling makes the information meaningful. They further noted that stories have a common language or taxonomy. Within a KMS, the taxonomy should be KM specific and that, in general, it should be industry immaterial, i.e. the story in one industry can be applied with success to another industry within the same CoP.

Figure 1 - Information systems are interrelated systems of technical and organizational elements

Success was measured against whether or not domain experts were willing to contribute their tacit knowledge to a KMS developed for the study.

It is acknowledged that not all tacit knowledge is best collected through an indirect knowledge transfer approach such as the implemented storytelling-based approach. Examples where this indirect knowledge transfer approach can be successful includes IT, law enforcement, and many others. This study concentrated on one CoP -

Hardware Software Data Processes People


20 approach, will achieve greater collection of viable tacit knowledge into a KMS.

Problem Statement

Domain experts (DE) are willing to share their personal (tacit) knowledge with others using direct knowledge transfer to help them learn but they are less willing to provide the same knowledge into a Knowledge Management Systems (KMS).

The research of Chun, Sohn, Arling, and Granados (2008) at Pratt-Whitney

Rocketdyne supports the willingness of DE to share their knowledge with others either by being asked or being presented with the opportunity to showcase and share their

knowledge. The Pratt-Whitney Rocketdyne KMS consisted of two components – the first was ‘Askme’ that consisted of chats, blogs, a forum, and ‘Goldfire’ that was an advanced search engine to perform searches across the company’s numerous sources. The research of Ko, Kirsch, and King (2005) further supports the willingness of DE to

transfer tacit knowledge. In the research of Ko, et al., the research was in the

transference of tacit knowledge between consultants and clients in Enterprise Resource Planning (ERP) systems implementation – not into a KMS. Wasko and Faraj (2005) noted that in Communities of Practices (CoP), a field of shared interests, knowledge flows easily within that CoP and enables the participants to create social networks to support the exchange of knowledge. In electronic CoPs, such as the ‘Goldfire’ forum, knowledge participants had no assurances that the people they were helping would reciprocate in kind when the participant needed help.

The research of Kalid and Mahmud (2008) concentrated on capturing tacit knowledge through stories with videos being the end result. They recognized that


21 transfer tacit knowledge (direct knowledge transfer) but they were not being captured into a KMS (indirect knowledge transfer). Kalid and Mahmud looked at storytelling from the perspective of verbal descriptions of information. According to Pumareja and Sikkel (2005), applications can be designed to collect tacit knowledge; however, if domain experts are unwilling to contribute, the KMS will fail. Weber (2007) went further and provided several reasons why KMS' fail:

• KMS' often fail due to the nature of the KMS, i.e. there are no bounds on what a domain expert can enter or how. Haller and Abecker (2010) considered the reliance on highly structured semantic meta data as a major challenge for KMS'.

• KMS’ may fail when they do not integrate humans, processes, and technology. This is justified by the limitations and importance of each of these components (Abecker, S., & Maurer, 2000).

• KMS’ may fail when users are afraid of the consequences of their contributions; this is related to job security. Users may also feel that withholding their knowledge may be a way to secure influence (Disterer, 2001).

Orth, Smolnik and Jennex (2009) described the different KMS approaches as IT-based systems that combine content, organizational processes, users and technical solutions which supports Weber (2007) who stated that KMS' fail when humans and processes are not integrated with technology. This also supports Riege (2005) who identified three key barriers to knowledge share – individual, organizational, and technology.


22 repository-based KMS' that utilized database management systems with data in a variety of formats. They noted that KMS information was often difficult to find, was not vetted prior to being made available to users and users found it difficult to relate the information to their problems.

Ariffin, et al. stated that it was difficult to motivate users to contribute their tacit knowledge to a KMS while Fanfan's (2012) research indicated that tacit knowledge could not be stored using a KMS. The issue of motivation was also noted by Sandhu, Jain, and Ahmad (2011) and Okoroji, Velu, and Sekaran (2014). Yao, Kam, and Chan (2007) also noted the issue of motivation, i.e. lack of motivation and/or reward for employees. Smuts, Merwe, Loock, and Kotze (2009) considered that while the collection of tacit knowledge is difficult to codify, tacit knowledge could be used to create new explicit knowledge; however, their research indicated that tacit knowledge cannot be easily articulated which corresponds to Fanfan (2012) and Patalas-Maliszewska, Krebs, and Dudek (2017).

The research into KMS’ and tacit knowledge has clearly shown that the reasons for the inability of KMS’ to capture tacit knowledge are still largely unknown and, in essence, marked by a high degree of variation. Some research such as Weber’s (2007) considers the KMS as the primary issue (technology). The majority of Weber’s issues dealt with the KMS such as how data was entered and the type of KMS framework. Others considered failure of KMS’ as more people-centric, i.e. why people do not want to enter or cannot enter data (Ariffin et al., 2007; Disterer, 2001; Fanfan, 2012; Smuts et al., 2009); this spans the individual and organizational barriers noted by Riege. People issues


23 include the lack of a CoP, the difficulty in codifying tacit knowledge, and motivating DE to contribute their tacit knowledge.

a KMS framework was developed for this study that addressed the three issues identified by Weber utilizing a storytelling approach. The study also addressed Riege’s issues of technology and people using a storytelling approach that was expected to motivate DE to contribute their tacit knowledge.

Research Questions

1. Proposition 1: Domain experts will be willing to provide their tacit knowledge into a KMS using a storytelling-based approach. The storytelling approach is how tacit knowledge is elicited from domain experts and how that knowledge is then communicated to users of the KMS. Instead of a simple fact-based

approach, a story is created that is more interesting to both the participant and the reader of the story who is searching for information. This proposition goes to the research of Schank (2010) and Whyte and Classen (2012) who noted that telling a story is more interesting than just static dictation, Qiu, et al. (2014) who stated that tacit knowledge cannot be codified and can only be observed and Rumanti, Hidayat, and Saputro (2015) who considered tacit knowledge to be one of the most crucial factors in small and medium enterprises yet also considered the most difficult to transfer to others. As noted earlier, the issues with tacit knowledge are not with direct knowledge transfer but with indirect knowledge transfer.

2. Proposition 2: The use of scenarios, defined CoPs, domains and subdomains, and guiding questions in a semi-structured format will resolve the issue that KMS' often fail due to the nature of the KMS, i.e. there are no bounds on what a domain expert can enter or how. The semi-structured format is one where the participant is free to tell their story as they feel it should be told; however, a structure exists


24 provide clarity and to reduce or eliminate assumptions the participant may make about the reader’s knowledge. Domains and subdomains provide for easy retrieval by those who seek the information. The CoP aspect was designed to address the issue of Weber (2007) that KMS’ fail because they are not designed to support communities of practice and Orth, et al. (2009) whose approach, based on processes or tasks (KM in the small), concentrated on employee usage of

knowledge in a task, process, or project that already possessed a common context of understanding, i.e. a CoP.

3. Proposition 3: People are more willing to provide their tacit knowledge in forums versus provide their tacit knowledge in corporate KMS'. The literature clearly demonstrates that people do not like to provide their tacit knowledge into

corporate KMS’; however, the literature also clearly demonstrates that people do provide their tacit knowledge into forums and like it. The Goldfire KMS has had some success (Chun et al., 2008). Goldfire was essentially a two-part

implementation where the first part, the AskMe portion, was fundamentally a forum. This proposition sought to identify a motivation for why participants are willing to provide their tacit knowledge to forums but not KMS’.

Relevance and Significance

The key relevance of this research is that the participants enter stories directly into the KMS themselves unlike prior research using stories where tacit knowledge was acquired from interviews and videos and then entered into the KMS by the interviewers. The storytelling approach of Azudin, Ismail, and Taherali (2009) documented the use of storytelling at lunch and other methods such as forums to enhance knowledge sharing and


25 collaboration. Furthermore, this research utilized domains and subdomains to catalog the stories into CoPs making it easier for those requiring the knowledge to find it.

The use of stories in this context is a variation of an approach that has demonstrated great success through interpersonal communications but has not been incorporated successfully into a KMS where the DE contribute their tacit knowledge directly into the KMS.

For this study, tacit knowledge from DE was collected as stories based upon scenarios under a relevant domain and subdomain within a CoP. An example of a domain/subdomain within the IT CoP and this study was provisioning hardware or software (domain) and servers (subdomain), i.e. hardware/software provisioning or the provisioning of hardware and/or software. In this way, not only did the tacit knowledge get stored but it was a valid mechanism for other DE to, in essence, peer review the input knowledge for relevancy and currency and to evaluate other potential

domains/subdomains where the knowledge might be applicable. Using this approach, it was believed by this researcher that users would be able to find relevant data quickly in much the same way that forums use catalogs such as Cameras/Canon/Bodies or

Cameras/Canon/Lens. The solution set encompassed the technology, the processes and the people as noted in Weber (2007) and covered the content as noted in Orth, Smolnik and Jennex (2009). For this study, domains/subdomains went no further than two levels deep although in a real system they could go four or more levels deep.

While the use of storytelling within the context of knowledge management to capture tacit knowledge is not unique, the use of storytelling as a means of populating a KMS by the holders of that knowledge directly into the KMS is. This study captured participant knowledge as well as lessons learned that were related to specific issues


26 and went well.

The significance of this research was the demonstration that relevant tacit

knowledge from domain experts could be acquired without the long, drawn out approach of interviews and videos that is time consuming and difficult to ensure currency. Thus, this research extended the storytelling approach from one-on-one interviews to direct input by the holders of the tacit knowledge using stories. The use of tacit knowledge through indirect communications has been used in forums and wikis.

This study provides researchers and implementers of KMS' another approach to capturing and making available to users the individual (tacit) knowledge of DE. Lastly, this study dispelled past research that stated that tacit knowledge cannot be successfully entered into KMS'.

Barriers and Issues

Several barriers affected this study. These are identified below along with resolutions:

1. As will be presented in the ‘Approach’ section, the creation of stories involves the use of multiple tables that break a story down into small bytes of data that, in the end, are consolidated into the creation of a single story. A story

consists of two major components - freeform data and response data with both being entered by the DE (participant). Response data addresses guiding questions. The guiding questions are not the story but provide a framework to elicit specific information regarding the story that is often lost due to

assumptions on the part of DE. This will be discussed in greater detail under the approach section. The challenge was the parsing of the input data into a


27 story that would be understood by others. As will be discussed in Chapter 3, this issue was resolved through formatting within the framework.

2. As noted in the first part of the literature review, storytelling has, in the past, been a video or face-to-face action while in this study, data was entered directly into the KMS by the participant. Thus, a key challenge was the development of a KMS with the necessary framework and processes to

successfully elicit a story related to a point the participant was trying to make. As noted in Kalid and Mahmud (2008), the technical language used by

contributors might not be understood by everyone within the same CoP. This could be especially problematic when the story is read by non-technical readers. Thus, the framework of the CoP, domains and subdomains must be to a level generally understood by the intended audience. Resolution was through use of the CoP, guiding questions, and the domains/subdomains and integration into the final story.

3. Once the KMS was developed and internally tested, participants needed to be identified to participate in the study. Finding participants was a major

challenge. Initially, it was thought that participants would come from large companies with large Information Technology (IT) divisions. Thus, packets were prepared and sent to five major technology firms. This proved less than satisfactory in obtaining participants. This barrier was overcome by soliciting participants on university web sites, knowledge management forums, and technical organizations in Facebook. While the desire was for the majority of participants to have some prior KMS knowledge enabling the comparison of this approach with past approaches, it was determined that a comparison of the approach itself could be done with participants who did not have prior


28 experience may be reluctant to participate for the same reasons they do (or did) not want to participate in their company KMS’. These reasons go back to the cultural barriers noted by Disterer (2001) and Weber (2007). Disterer noted that cultural traditions tend to discourage knowledge sharing. One cultural barrier he noted was ‘knowledge is power’ where experts with rare knowledge results in knowledge hoarding instead of transfer. McDermott and O’Dell (2001), however, believed that such barriers could be overcome. McDermott and O’Dell felt that you do not change the culture to match the knowledge management initiatives but adapt the approach to knowledge management to fit the culture. This was the approach taken in this study. 4. The framework would be a critical factor to success. The framework defines

how a story is told. It defines the various attributes that ultimately lead to a complete story. Telling a story as a scenario that is not filled with useless information ("I started my day off with a bowl of cereal, I really like cereal and then...") or making assumptions about reader knowledge, even within a CoP, was challenging.

Two paths were considered – written and audio. Each has important arguments. Audio is more natural and reduces the overall number of

attributes (columns) within the story table for stories-in-development;

however, the audio requires storing files to disk increasing complexity and has the potential of reducing anonymity because someone may recognize the voice. Audio also injects significant challenges if the participant wishes to make changes to the story. Writing is more time consuming and stories-in-development must be broken down into more, and smaller, attributes to assist


29 in the development of the story. Then there is the challenge of participants being poor writers. On the other hand, making changes to a story is relatively easy. In the end, writing was chosen as the mechanism for the framework as it is easier to control the creation of the story.

Assumptions, Limitations and Delimitations

• It was assumed that participants who like telling stories would have a higher number of contributions to the KMS than those who do not like telling stories. This

assumption was not validated as no participant entered more than one story.


• The literature review will show that the issues of tacit knowledge are global. This research, however, relied solely on participants located within the United States. While some participants may originate from other countries, it is likely that most participants will be American or have considerable time within the United States. For clarity in stories, all participants were required to write in English. Global cultural differences could affect how participants interact with the KMS thus potentially altering the results positively or negatively.

• This research conducted a limited comparison to social media data collection, i.e. forums. However, an in-depth analysis of the differences was beyond the scope of the study. This topic is covered in more detail in Chapter 3, under ‘Design of the KMS’.


• The study focused on the area of IT. There were two primary reasons for focusing on IT: my primary skill sets are in IT and, IT is an area rich in communities of practice.


30 analysis of the data. Thus, for the purpose of this study, IT was the high-level CoP. • Within IT, the study focused on a limited number of domains and subdomains. The

chosen domains and subdomains are relatively common within IT and thus aided in obtaining participants for the study.

• In the study, collected data relating to social media forums was considered ‘micro’ knowledge vs. ‘macro’ knowledge used in traditional KMS’. In this way,

comparisons between forums and traditional KMS’ became clearer. Definitions for ‘micro’ and ‘macro’ knowledge are provided in the ‘Definition of Terms’.

Definition of Terms

Case-Based-Reasoning (CBR) – Reasoning by analogy

Community of Practice (CoP) - The members share the same interests within the community.

Content Management Systems (CMS) - A content management system is software or a group or suite of applications and tools that enable an organization to seamlessly create, edit, review and publish electronic text.

Database Management Systems (DBMS) – Software that handles the storage, retrieval, and updating of data in a computer system (Dictionaries, 2017)

Domain Experts (DE) – the same as subject matter experts (SME), i.e. individuals who are considered experts with expertise in one or more areas.

Domain - In the context of this study, a primary category denoting some activity such as migrating something or provisioning something.


31 Enhanced Entity Relationship (EER) Model - Provides the detailed view of the table structure for the KMS

E-Learning - Electronic learning

Explicit Knowledge - Explicit knowledge is that knowledge that can be expressed in words, numbers, and symbols, and stored. It is knowledge that is recorded and easily expressed.

FOC - Failover Cluster – where two servers exist with one being the primary and the other being the backup; both servers are always running and failover is automatic.

IP - The Internet Protocol (IP) is the principal communications protocol in the Internet protocol suite for relaying datagrams across network boundaries (Tanenbaum & Wetherall, 2011).

IRB - Institutional Review Board – required when human subjects are used in a study.

Knowledge Management (KM) in the large - An approach based on infrastructure or generic systems (KM in the large) - concentrates on usage of knowledge where users do not have a common context of understanding (not a CoP).

KM in the small - An approach based on processes or tasks (KM in the small) - concentrates on employee usage of knowledge in a task, process, or project that already possesses a common context of understanding (a CoP).

KMS - Knowledge Management System, i.e. the technology used to implement KM.

LB - Load-Balanced (usually for LB clusters for use in webs). A LB cluster balances users across n number of web servers thus ensuring an even load. If a server


32 anticipated load.

Macro Knowledge - Knowledge that typically exists in KMS’. Reusable knowledge that is likely to be needed by others.

Micro Knowledge - Knowledge that typically exists in forums and other social media. Knowledge needed by one person for a specific task that is likely to be discarded knowledge.

Datacenter Migration - Migration of data from one data center to another. An example is closing a data center and migrating hardware and software to another data center.

Application Migration - Migration of an application from test into production or to another server; includes software updates and patches.

Ontology - In the context of computer and information sciences, an ontology defines a set of representational primitives with which to model a domain of knowledge (Liu & Ozsu, 2009)

Scenario - A scenario is a situation in which a story takes place. It could be a problem that occurred and was resolved or lessons learned as a result of a well-planned activity.

Story - The real-life experiences and lessons learned of domain experts in their area of expertise.

o Characters - Individuals involved in the story o Plot - The problem or solution


33 o Setting - Where, and when

o Theme - How, why, conflicts encountered, and lessons learned

Subdomain - A sub category of a domain. Within IT, if a domain is migration, a subdomain could be servers, i.e. the migration of servers. Subdomains are merely subcategories of the higher level, i.e. domains.

Tacit Knowledge - Tacit knowledge, according to the Cambridge Dictionary, is knowledge you get from personal experience. The Law Dictionary states that tacit

knowledge is unspoken, unwritten, and hidden stores of knowledge based on experiences, emotions, institutions, insights, and observations.

Virtualization - Virtualization is an abstraction of the logical to the physical. An example is a virtual machine which is not a physical system but logical within a physical entity. Virtualization can encompass servers and operating systems, storage, and



Traditional KMS’, while experiencing some successes (mostly with explicit knowledge), have, for the most part, been unsuccessful when it comes to capturing the tacit knowledge of domain experts. Several reasons have been put forth in the research as to why this is such as knowledge is power, and poor technology.

For this study, a limited-use KMS was developed that facilitated the input of tacit knowledge through storytelling, implemented guiding questions to reduce the likelihood of assumptions, and operated within a community of practice. The structure of the knowledge management system was built around a framework designed to elicit tacit knowledge through a semi-structured approach using narrative stories.


34 through guiding questions; the framework and guiding questions are covered in-depth in Chapter 3, Methodology. The framework was the fundamental mechanism that

ultimately resulted in a story that was readable, and correctly covered a scenario such that the reader gained what they desired from the story. Thus, it was critical to success. The specifics of the framework followed characteristics of a narrative story discussed in Linde (2001).

In general, people like to tell stories. Many cultures have used stories to pass on their histories and lessons from generation-to-generation. Many domain experts use stories to pass on information to other experts in order to get a point across. They often use stories as a way to communicate critical issues and lessons learned to more junior members (direct communications) in order to help them learn their trade or to

management to help them understand the issues.

Research on telling stories has been through direct communications, i.e. person-to-person as noted above or through interviews. After conducting a literature search of over 100 articles spanning journals and conference proceedings, evidence strongly supports that this study was unique by extending the use of stories from a direct

knowledge transfer mechanism to an indirect knowledge transfer mechanism where the DE entered the story themselves into the KMS without an intermediary element.

This study looked at two methods for participants telling their stories. The first was audio and the second written text. Each had specific advantages and disadvantages. This study utilized written text. While people may not write well, much of the poor writing can be corrected and people, generally, will still understand a story even if it is poorly written. On the other hand, people have a very difficult time understanding a


35 story where they do not understand the speaker and the differences can be as simple as the listener being from one part of the country while the speaker is from another part of the same country. Lastly, written text was easier to control in the development of a story and much easier to make changes to.



Chapter 2: Review of the Literature

Knowledge Sharing

There is a good deal of research into the area of knowledge sharing which includes the sharing of tacit knowledge through direct (one-to-one, one-to-many, and many-to-many) communications. Where the research fails is in the area of indirect communications (such as a KMS). Yao, Kam, and Chan (2007) investigated how culture, attitudes, and barriers affect knowledge sharing in a Hong Kong government department. Forty people responded to their surveys about organizational culture and individual approaches towards knowledge. Seventy-nine percent (79%) either agreed or strongly agreed that knowledge was power. This can lead to knowledge hoarding but 95%

indicated that they liked to share knowledge. Over seventy present (70%) felt that lack of incentives/rewards, lack of time, and a weak culture of sharing were barriers to

knowledge sharing. Okoroji, Velu, and Sekaran (2014) also found that appropriate motivation of employees is important for a successful knowledge sharing process. They found that the voluntary nature of knowledge sharing participation retards efforts of most organizations towards effective KM. The central finding of McDermott and O’Dell (2001) was that an organization may have a strong commitment and approach to KM; however, the KM approach must accommodate the organization’s culture instead of trying to change the culture to fit the approach. In their research, stories also factored in, i.e. stories were used to communicate what attitudes and actions were acceptable and unacceptable. Saenz (2012) concluded that knowledge sharing is key to innovation and that sharing can come from on-line discussion forums, blogs, intranets and knowledge


37 repositories; however, personal interaction between individuals is another mechanism for sharing tacit knowledge such as in CoPs, coaching, and mentoring.

Knowledge Management Systems

Knowledge Management (KM) addresses the process of acquiring, creating, distributing and using knowledge in organizations and knowledge management systems (KMS) are the technological mechanism for implementing knowledge management (Becerra-Fernandez, 2000; Frost, 2013; Jimenez-Jimenez et al., 2014; Rance & Hanna, 2007) .

Chatti, Schroeder, and Jarke (2012), in their research into KM and Technology-Enhanced Learning (TEL), stated that since the introduction of KMs in the early 1990s, KMS's have failed to address the challenge of increasing knowledge worker productivity due to rapid changes in knowledge. The rapid change in knowledge is one of the reasons that the collection of tacit knowledge through interviews and videos is impractical. Knowledge is the primary resource for individuals. They attributed the challenges to different and incompatible concepts and tools for KM and TEL. They also highlighted that the relationship between KM and TEL needs to be closer. Kulkarni, Ravindran and Freeze (2007) stated that knowledge consists of explicit and tacit knowledge; both cannot be managed in the same manner. Their research indicated that the strategy for knowledge transfer of tacit knowledge is direct contact such as apprenticeship and mentoring;

however, their research focused on explicit knowledge.

Thalmann, et al. (2010) considered the variety of knowledge work environments at different organizations and considered the many environments as a negative factor affecting productivity, i.e. the lack of standards in KM while Dingsoyr, Djarraya, and Royrvik (2005) looked at how existing tools were being used in organizations. They


38 identified two strategies - those that focus on codifying relevant knowledge (technology) and those that count on communications between people with relevant tacit or explicit knowledge (personalization). Their research looked at tools to enhance personalization and found that many companies had developed KM tools to survey what type of

knowledge people had and then index it. The indexing was a use of ontologies. The development of ontology's was supported by Lee (2012) whose research concluded that accessing the appropriate knowledge can be difficult, time consuming and frustrating. His research showed that many organizations suffer not from the lack of knowledge but from ways of accessing and exploiting existing knowledge. Wu (2008) also looked at ontology models that identified documents and other explicit and tacit knowledge, i.e. how to find the knowledge. His research, like Lee, looked at maps to show where the knowledge is. Abdullah , Eri, and Talib (2011), while not specifically addressing explicit or tacit knowledge, discussed the importance of CoPs. Their main contribution was to propose a model to manage and facilitate CoP knowledge using KMS techniques;

Abdullah, et al. used Yellow Pages as an example (explicit knowledge). Makolm, Weiss and Reisinger (2007) found that knowledge workers require a certain degree of freedom in structuring their own tasks which often conflicts with the organization's needs for standardization.

Yordanova (2007) looked at common features of KM and E-learning. In Content Management Systems (CMS), the author described Learning Objects (LOs) that were used for presentation of learning content and knowledge. LOs, small independent units of information that could be combined in different contexts, were used for development and exchange of different types of information. Marshall, et al. (2003) developed a system, GetSmart, designed to apply KM techniques and integrate search tools with


39 concept mapping. The goal was preservation of data. Similar to Schank (2010), Eales (2004) sought a different approach. His approach looked at situated learning potential from the perspective of collaborative support provided by colleagues. He argued that we need to move beyond knowledge management and instead move to sharing expertise. Guechtouli (2012) looked at transference of tacit knowledge from experts to newcomers needing the knowledge. She investigated the use of CoP's and concluded that the impact of communicating knowledge is based on how the recipient views the contributor who is providing the knowledge. Guechtouli also noted two different types of knowledge transfer - direct and indirect. Direct transfer correlates to personalization (person-to-person) while indirect transfer correlates to persistent mechanisms, i.e. forums, wikis, and other similar methods. Her research supported that indirect communication enables more powerful knowledge transfer and can be used by different people which increases the ability of the knowledge to spread. Purcell and O’Brien (2015) noted, like others, that tacit knowledge is aligned with competitive advantage. Khan, Prasad, Selvi, et al. (2015) noted that tacit knowledge is difficult to capture or share while Khalid, Shehryar, and Arshad (2015) stated that tacit knowledge cannot be shaped and transported between organizations because of cultural, structural, and goal differences.

Xinxiang and Xiaohui (2011) noted that in a Delphi Group survey, 42% of respondents considered tacit knowledge more important than explicit knowledge and one of the goals of KM to be the transfer of individual (tacit) knowledge into group (explicit) knowledge. Chen, Xiao, Ren and Shi (2011) did not consider knowledge sharing as the ultimate goal of KM, simply a means. They understood that the acquisition of tacit knowledge is not simple and requires a comprehensive extraction process. Thus, their research goal was to eliminate the obstacles of knowledge exchange. Hsu and Sabherwal


40 (2011) contended that academia and practitioners consider the importance of intellectual capital (tacit knowledge) as a major source of sustained competitive advantage. Thus, they considered the externalization of one's tacit knowledge into a KMS a major issue.


The definition of a forum chosen for this study was given in Morzy (2010) when he wrote that an Internet forum is a Web application for publishing user-generated content under the form of a discussion. Usually, the term forum refers to the entire community of users with discussions on particular subjects called topics or threads. Posted messages are displayed chronologically (topics or threads).

Cerulo and Distante (2013) noted that even with forums that are organized and moderated by topics, discussions ‘tended’ to host messages on related subjects while Morzy’s (2010) research showed that discussions on forums are often shallow, emotional, inconsistent, lacking discipline and manner; they rarely contain useful practical

knowledge or specialized information. Sani, Kardan, and Cohan (2013) concluded that due to the large amount of information in forums, finding appropriate answers is becoming more time consuming and there is no suitable mechanism to measure the reliability of the answers being provided. Wasko and Faraj (2005) stated similarly when their research found that those seeking knowledge have no control over who responds to their questions or the quality of the responses. Sani, et al. also concluded that search engines are unable to process queries to questions. Ni and Li (2012) found that in online forums, a user’s interests are reflected via the contents generated by them, the users they exchange opinions with and the topics of discussions they participate in.



KMS Approaches

Orth, Smolnik and Jennex (2009) describe the different KMS approaches as IT-based systems that combine content, organizational processes, users, and technical solutions. They describe the types of implementations as:

• Approaches that are based on infrastructure or generic systems (KM in the large) – they concentrate on employee usage of knowledge where users do not have a common context of understanding (not a CoP).

• Approaches that are based on processes or tasks (KM in the small) – they concentrate on employee usage of knowledge in a task, process, or project that already possesses a common context of understanding (a CoP).

• Integrated approaches which attempt to combine both KM in the large and KM in the small.

KMS Architectures

Different architectures have also been proposed for KMS' that include Database Management Systems (DBMS), Case-based Reasoning (CBR) and ontology's.

Database Management Systems

Weber and Gunawardena (2008) and Benbya and Alstyne (2008) looked at repository-based KMS' that utilized database management systems with data in a variety of formats. Repository-based KMS' are used for knowledge sharing and leveraging of knowledge. Both papers noted that information was difficult to find, was not vetted prior to being made available and users found it difficult to relate the knowledge to solving their problems.



Case-Based Reasoning (CBR)

Maalel, Mejri, Mabrouk, and Ghezela (2012) and Weber and Gunawardena (2008) looked at CBR. CBR is applying past situations that are similar to a current situation to help resolve the current situation; CBR is a form of reasoning by analogy. Three types of CBR knowledge were discussed:

• Procedural - how a problem may be solved

• Declarative - what is known about a problem

• Heuristic - knowledge usually discovered through experience that has specific applicability (tacit knowledge)

Ontology-based KMS’

The research of Maalel, et al. (2012) considered that ontology-based KMS' could significantly reduce the effort of acquiring knowledge and could help to establish a common vocabulary for describing a situation and be used to model the knowledge necessary for indexing and organizing events. An ontology-based KMS uses a rigid structure based upon a library of keywords. Nasir and Noor (2010) developed an ontology-based KMS approach for e-applications on the web. Basically, ontology takes knowledge into another level where it gives meaning to content. This fits well with Chakraborty, Nayek, Basak, Ghosh and Debnath (2010) who saw a KMS as a simple query-response model used to extract tacit knowledge. Chakraborty, et al. saw an ontology-based KMS being faster than a DB-based KMS.


Sole and Wilson (2002) stated in their Harvard paper that organizations and their leaders are paying increasing attention to the role and value of narrative and anecdotal


43 information conveyed in the form of stories. They stated that knowledge cannot be completely abstracted into categorical and analytical forms and is inadequately conveyed in such forms. Schank (2010) discussed the art and importance of storytelling and tied that to the actions of a company that collected video stories of their best people. Some of the stories were applicable to specific situations and Schank was working to index the stories to enable employees to find the video stories when doing something specific. Schank (2010) observed that in the age of the Internet, companies have too many electronic documents. Due to the large volume of documents, many e-mail recipients often do not open their files. The large volume of electronic data contributes to

challenges of KMS'. Schank further observed that before the Internet, knowledge was passed on by stories. Whyte and Classen (2012) also researched the use of storytelling to elicit tacit knowledge from subject matter experts (SME). Whyte and Classen collected their story data through one-on-one interviews and felt that stories make information meaningful and are the best way to transfer tacit knowledge. They collected their information through interviews using guided questions (not to be confused with guiding questions in this study); SMEs were presented with a prompt card containing a brief array of story types to help them recall stories. Their intent was to identify a common language or taxonomy, identify a taxonomy that was KM specific and that was not specific to any industry. In the 1990s, Xerox field employees, through direct communications, were found to be passing on their tacit knowledge at water coolers on how to repair equipment better. Thus, the tech reps went from being independent workers to social learning units (Sole & Wilson, 2002). Azudin, Ismail, and Taherali (2009) researched knowledge